The Claude Code Blueprint.

Get Claude Code installed and shipping in 60 minutes.

Free · 60 seconds to opt in

What you walk away with.
  • A working /brief command running three research subagents in parallel
  • Real company briefs landing in your own Airtable
  • A live URL on Vercel — actual proof you can do this
  • Five short interactive lessons. You build, I guide.
tom — claude-code — 80×24
~/briefs ❯ claude
Welcome back, Tom. Ready to build.
 
claude ❯ /brief acme.com
↳ Dispatching 3 subagents in parallel...
Company overview [done]
Recent news (12mo) [done]
Key people + roles [done]
 
Brief saved to Airtable.
Live URL: briefs.tomcrawshaw.dev/acme
 
// 60 seconds. From your terminal.
// Built inside Claude Code.
 
claude ❯

The watchers fall behind.

They bookmark threads, save demos, screenshot prompts. Six months in, they've consumed hundreds of hours of AI content and shipped exactly nothing. The space is moving. They're not.

You're not behind because the tools are too hard. You're behind because nobody's told you what to build first. So you keep watching, and the gap widens, and the people in your shoes who started building three months ago are eating your lunch.

You don't have a knowledge problem. You have a direction problem.

Architects don't need another course. They need someone telling them what to build first, and sitting beside them while they build it. That's all this is.

The toolchain is Claude Code from Anthropic, and n8n. Both are stable, well-documented, and the right substrate to build on for the next five years.

— Tom Crawshaw
If you've already heard this and you're still watching —

Architects ship.

Three ways in.

Pick the one
that fits.
Tom Crawshaw, founder of AI Architects
From the builder

Tom Crawshaw.

Eight years building automations. Over $25 million in revenue generated for clients using AI and workflow systems across private aviation, construction, marketing agencies, and e-commerce.

I don't sell information — YouTube has plenty. What it doesn't have is someone looking at your specific setup, telling you exactly what's wrong, and getting you unstuck in the same session.

$25M+
Generated for clients
8 yrs
Building automations

Why AI Architects.

What makes
this different.

An operator, not a course-seller.

Eight years building automations and over $25 million generated for real clients. This is built on client work, not a content channel.

You ship live, not watch.

Every path ends with a working agent and a live URL on Vercel. You walk away with something running, not a certificate.

Your business, on the screen.

We build against your actual bottlenecks, not a generic curriculum. The work is your work, from the first session.

See what operators have built inside their businesses.Read the case studies
Common questions

Before you commit.

What is Claude Code, and do I need to know how to code to use it?

Claude Code is Anthropic's command-line coding agent. You type plain English; it edits files, runs commands, and ships projects in your terminal. You don't need a CS degree — most mentorship clients had never touched a terminal before week one. You do need to be willing to read what the agent is doing and override it when it goes sideways.

How is this different from a YouTube channel or a $97 course?

Courses sell information. Mentorship sells direction. The bottleneck is not 'what to learn next' — every Claude Code tutorial is on YouTube already. The bottleneck is 'what should I build first, in my business, with my data, this Wednesday?' I sit on the call and answer that, then watch you build it.

How long does it take to ship something real?

The free Blueprint puts a working agent and a live URL in front of you in 60 minutes. The 30-day Challenge ships four working automations and a deployed app in four weeks of 90-minute weekly calls. The mentorship is 90 days, by application, for operators who want a trusted advisor while they re-platform their business onto AI.

I'm already using ChatGPT or Cursor — is Claude Code worth switching to?

Cursor edits code inside an IDE. Claude Code runs in your terminal and executes commands, edits files across a project, and orchestrates subagents in parallel. They solve different problems. Most students run both. For operators automating their own business — not writing app code — Claude Code wins on subagents, skills, and CLAUDE.md project memory.

How much does Claude Code itself cost on top of the program?

Anthropic's Pro plan is $20/month and covers the daily work for most operators. Heavy users move to Max ($100 or $200/month). API pricing is separate and only matters if you're hammering it with batch jobs. None of the programs here require a specific tier — the Blueprint runs on the free Anthropic account.

Can I talk to someone before I commit to the paid programs?

Yes. Start with the free Blueprint to confirm Claude Code clicks for you. If you want the 30-Day Challenge or 90-day mentorship, the next step is an application call. No pressure pitch — if it's not a fit I'll point you at a cheaper path.

"You don't need a course. You need one good Wednesday."

Stop bookmarking.
Start shipping.